Calcium‐based stabilizer materials (CSMs) exhibit pozzolanic properties which improve the properties of clayey soils by hydration, cation exchange, flocculation, pozzolanic reaction, and carbonation. In this comprehensive review, comprising over past three decades from 1990 to 2019, a mechanistic literature of expansive soil stabilization by incorporating CSMs is presented by reviewing 183 published research articles. The advantages and disadvantages of CSMs as the ground stabilizing agent are succinctly presented, and the major outcomes of physicochemical effects on soil properties are discussed in detail. After blending with CSM, the main and interaction effects on soil properties with focus on chemical processes such as X‐ray fluorescence, X‐ray diffraction analyses, and microstructure interaction by using scanning electron microscopy and thermogravimetric analysis have been reviewed in light of findings of past researchers. This work will help geotechnical engineers to opt for suitable CSM in the field of geoenvironmental engineering in committing to sustainable construction of civil engineering structures over expansive soils.

This research suggests a robust integration of artificial neural networks (ANN) for predicting swell pressure and the unconfined compression strength of expansive soils (P_sUCS-ES). Four novel ANN-based models, namely ANN-PSO (i.e., particle swarm optimization), ANN-GWO (i.e., grey wolf optimization), ANN-SMA (i.e., slime mould algorithm) alongside ANN-MPA (i.e., marine predators’ algorithm) were deployed to assess the P_sUCS-ES. The models were trained using the nine most influential parameters affecting P_sUCS-ES, collected from a broader range of 145 published papers. The observed results were compared with the predictions made by the ANN-based metaheuristics models. The efficacy of all these formulated models was evaluated by utilizing mean absolute error (MAE), Nash–Sutcliffe (NS) efficiency, performance index ρ, regression coefficient (R²), root mean square error (RMSE), ratio of RMSE to standard deviation of actual observations (RSR), variance account for (VAF), Willmott’s index of agreement (WI), and weighted mean absolute percentage error (WMAPE). All the developed models for P_s-ES had an R significantly > 0.8 for the overall dataset. However, ANN-MPA excelled in yielding high R values for training dataset (TrD), testing dataset (TsD), and validation dataset (VdD). This model also exhibited the lowest MAE of 5.63%, 5.68%, and 5.48% for TrD, TsD, and VdD, respectively. The results of the UCS model’s performance revealed that R exceeded 0.9 in the TrD. However, R decreased for TsD and VdD. Also, the ANN-MPA model yielded higher R values (0.89, 0.93, and 0.94) and comparatively low MAE values (5.11%, 5.67, and 3.61%) in the case of PSO, GWO, and SMA, respectively. The UCS models witnessed an overfitting problem because the aforementioned R values of the metaheuristics were 0.62, 0.56, and 0.58 (TsD), respectively. On the contrary, no significant observation was recorded in the VdD of UCS models. All the ANN-base models were also tested using the a-20 index. For all the formulated models, maximum points were recorded to lie within ± 20% error. The results of sensitivity as well as monotonicity analyses depicted trending results that corroborate the existing literature. Therefore, it can be inferred that the recently built swarm-based ANN models, particularly ANN-MPA, can solve the complexities of tuning the hyperparameters of the ANN-predicted P_sUCS-ES that can be replicated in practical scenarios of geoenvironmental engineering.

The California Bearing Ratio (CBR) is an important index for evaluating the bearing capacity of pavement subgrade materials. In this research, random subspace optimization-based hybrid computing models were trained and developed for the prediction of the CBR of soil. Three models were developed, namely reduced error pruning trees (REPTs), random subsurface-based REPT (RSS-REPT), and RSS-based extra tree (RSS-ET). An experimental database was compiled from a total of 214 soil samples, which were classified according to AASHTO M 145, and included 26 samples of A-2-6 (clayey gravel and sand soil), 3 samples of A-4 (silty soil), 89 samples of A-6 (clayey soil), and 96 samples of A-7-6 (clayey soil). All CBR tests were performed in soaked conditions. The input parameters of the models included the particle size distribution, gravel content (G), coarse sand content (CS), fine sand content (FS), silt clay content (SC), organic content (O), liquid limit (LL), plastic limit (PL), plasticity index (PI), optimum moisture content (OMC), and maximum dry density (MDD). The accuracy of the developed models was assessed using numerous performance indexes, such as the coefficient of determination, relative error, MAE, and RMSE. The results show that the highest prediction accuracy was obtained using the RSS-based extra tree optimization technique.

Coal mining waste in the form of coal gangue (CG) was established recently as a potential fill material in earthworks. To ascertain this potential, this study forecasts the strength and California Bearing Ratio (CBR) characteristics of chemically stabilized CG by deploying two widely used artificial intelligence approaches, i.e., artificial neural network (ANN) and random forest (RF) regression. In this research work, varied dosage levels of lime (2, 4, and 6%) and gypsum (0.5, 1, and 1.5%) were employed for determining the unconfined compression strength (UCS) and CBR of stabilized CG mixes. An experimental study comprising 384 datasets was conducted and the resulting database was used to develop the ANN and RF regression models. Lime content, gypsum dosage, and 28 d curing period were considered as three input attributes in obtaining three outputs (i.e., UCS, unsoaked CBR, and soaked CBR). While modelling with the ANN technique, different algorithms, hidden layers, and the number of neurons were studied while selecting the optimum model. In the case of RF regression modelling, optimal grid comprising maximal depth of tree, number of trees, confidence, random splits, enabled parallel execution, and guess subset ratio were investigated, alongside the variable number of folds, to obtain the best model. The optimum models obtained using the ANN approach manifested relatively better performance in terms of correlation coefficient values, equaling 0.993, 0.995, and 0.997 for UCS, unsoaked CBR and soaked CBR, respectively. Additionally, the MAE values were observed as 45.98 kPa, 1.41%, and 1.18% for UCS, unsoaked CBR, and soaked CBR, respectively. The models were also validated using 2-stage validation processes. In the first stage of validation of the model (using unseen 30% of the data), it was revealed that reliable performance of the models was attained, whereas in the second stage (parametric analysis), results were achieved which are corroborated with those in existing literature.

An accurate calculation of the flexural capacity of flexural members is vital for the safe and economical design of FRP reinforced structures. The existing empirical models are not accurately calculating the flexural capacity of beams and columns. This study investigated the estimation of the flexural capacity of beams using non-linear capabilities of two Artificial Intelligence (AI) models, namely Artificial neural network (ANN) and Random Forest (RF) Regression. The models were trained using optimized hyperparameters obtained from the trial-and-error method. The coefficient of correlation (R), Mean Absolute Error, and Root Mean Square Error (RMSE) were observed as 0.99, 5.67 kN-m, and 7.37 kN-m, for ANN, while 0.97, 7.63 kN-m, and 8.02 kN-m for RF regression model, respectively. Both models showed close agreement between experimental and predicted results; however, the ANN model showed superior accuracy and flexural strength performance. The parametric and sensitivity analysis of the ANN models showed that an increase in bottom reinforcement, width and depth of the beam, and increase in compressive strength increased the bending moment capacity of the beam, which shows the predictions by the model are corroborated with the literature. The sensitivity analysis showed that variation in bottom flexural reinforcement is the most influential parameter in yielding flexural capacity, followed by the overall depth and width of the beam. The change in elastic modulus and ultimate strength of FRP manifested the least importance in contributing flexural capacity.